Protective Effect of Pyrroloquinoline Quinine on Ultraviolet a Irradiation

4382 MOLECULAR MEDICINE REPORTS 12: 4382-4388, 2015

Protective effect of pyrroloquinoline quinine on ultraviolet A irradiation‑induced human dermal fibroblast senescence in vitro proceeds via the anti‑apoptotic sirtuin 1/nuclear factor‑derived erythroid 2‑related factor 2/heme oxygenase 1 pathway

CHUNLI ZHANG1, CHUANJUN WEN2, JINDE LIN3 and GAN SHEN3

1Department of Clinical Research, Friendship Plastic Surgery Hospital, Nanjing Medical University, Nanjing, Jiangsu 210029; 2College of Science, Nanjing Normal University, Nanjing, Jiangsu 210046; 3Department of Plastic Surgery, The Second Affiliated Hospital of Nanjing Medical University, Nanjing, Jiangsu 210000, P.R. China

Received May 20, 2014; Accepted February 13, 2015

DOI: 10.3892/mmr.2015.3990

Abstract. The aim of the present study was to determine blot analysis revealed significant effects of PQQ on SIRT1 whether pyrroloquinoline quinine (PQQ) exerts a protective and SIRT6. Nrf2 and HO‑1 exbibited mild changes with the effect on ultraviolet A (UVA) irradiation‑induced senescence same trend when treated with or without UVA and PQQ. in human dermal fibroblasts (HDFs) and to elucidate its mech- In conclusion, the results of the present study showed that anism of action in vitro. A senescence model was constructed pyrroloquinoline quinine may have a protective effect on UVA as follows: HDFs (1x104‑1x106) were cultured in a six‑well irradiation‑induced HDF aging, which may be associated plate in vitro and then exposed to UVA irradiation at a dosage with the anti‑apoptotic SIRT1/Nrf2/HO‑1 pathway as well as of 9 J/cm2. Various concentrations of PQQ (50, 100 and SIRT6 signaling. 200 ng/ml) were added to the culture medium 24 h prior to UVA exposure. Following 72 h of irradiation, senescence‑associated Introduction β‑galactosidase staining was performed in order to evaluate the senescence state. Furthermore, mRNA expression of the Skin is an essential natural barrier to protect body from outside senescence marker genes matrix‑metalloprotease (MMP)1 physical, chemical, microbial and other infestation; in addi- and MMP3 was determined using reverse transcription tion, skin is a visual indicator of the body's aging process (1). quantitative polymerase chain reaction. Protein expression 90% of the ultraviolet (UV) radiation reaching the surface of of sirtuin (SIRT)1, SIRT6, nuclear factor erythroid 2‑related the earth is long-wavelength radiation (UVA; 320‑400 nm), factor 2 (Nrf2) and heme oxygenase 1 (HO‑1) were detected which is 20 times higher than that of medium-wavelength using western blot analysis. The results showed that the radiation (UVB; 280‑320 nm), while short-wavelength radia- percentage of cells stained by X‑gal following 9 J/cm2 U VA tion (UVC; <290 nm) is completely absorbed by the ozone irradiation was markedly increased compared with that of the shield (2). Under physiological conditions, UVA penetrates the control group (53 and 8%, respectively), while 50 ng/ml PQQ epidermis into the dermis; therefore, it has been well‑estab- attenuated the ratio of positive staining compared with that lished that UVA irradiation is responsible for photoaging and of the UVA‑only cells (29 vs. 53%, respectively). Expression photocarcinogenesis (3). of fibroblast senescence marker genes MMP1 and MMP3 Pyrroloquinoline quinine (PQQ) is a non‑covalently was decreased in cells treated with UVA and 50 ng/ml PQQ bound redox co‑factor of bacterial dehydrogenases, which was compared with that of cells in the UVA‑only group. Western initially isolated from cultures of methylotropic bacteria, and does not rely on nicotinamide adenine dinucleotide phosphate (NADP) and NAD as well as flavin adenine dinucleotide (FAD). It was later reported that PQQ was abundant in various Correspondence to: Mr. Gan Shen, Department of Plastic Surgery, types of plant as well as in milk, animals and humans. PQQ The Second Affiliated Hospital of Nanjing Medical University, was therefore defined as a novel type of vitamin (4), which 262 North Zhongshan Road, Nanjing, Jiangsu 210000, P.R. China was found to have important roles in the promotion of growth, E‑mail: [email protected] cell proliferation and growth factor secretion (5). Evidence has suggested that PQQ is also involved in redox processes Key words: pyrroloquinoline quinine, ultraviolet A irradiation‑induced in the mitochondrial respiratory chain, scavenging of reac- senescence, sirtuin 1/nuclear factor‑derived erythroid 2‑related tive oxygen species (ROS), attenuation of oxidative stress in factor 2/heme oxygenase 1 pathway mitochondria (6) and the protection of neurons (7); in addition, PQQ was reported to antagonize several types of oxidative stress‑induced cell damage, including reoxygenation injury in ZHANG et al: EFFECT OF PQQ ON HDF SENESCENCE FROM UVA IRRADIATION AND THE ANTI-APOPTOTIC PATHWAY 4383 the heart, ethanol‑induced liver damage and hyperoxia‑induced the six‑well plate was placed on a brass block embedded on cognitive deficits (8). The present study aimed to determine ice, in order to reduce any evaporation, at a distance of 15 cm whether PQQ was involved in the protection of human dermal from the UVA light source. As the UVA irradiation source, fibroblasts (HDFs) from UVA‑induced senescence. an Ultraviolet phototherapy instrument (SS-04A; Shanghai In order to investigate the protective mechanisms of PQQ SIGMA High‑Tech Co., Ltd, Shanghai, China) equipped with a against UVA‑induced HDF aging, the present study assessed 15-W ozone‑free UVA lamp (CEL015 W; Philips, Groningen, the expression of two members of the sirtuin (SIRT) family: Holland) was used. The incidence dose of UVA was measured SIRT1 and SIRT6. Nuclear and cytoplasm‑localized SIRT1 with a UVA/UVB‑ultraviolet meter (Factory affiliated to has been reported to have important roles in apoptosis, differ- Beijing Normal University, Beijing, China). Following expo- entiation and oncogenic transformation (9‑14). SIRT1 was also sure to UVA for varying lengths of time, PBS was replaced found to deacetylate transcription factors and co‑activators, with culture medium and cells were incubated under standard such as heat shock factor 1, which induces the transcription conditions for 72 h prior to analysis. of molecular chaperones associated with the pathogenesis of Parkinson's disease and amyothrophic lateral sclerosis (15‑20). Total RNA isolation. UVA irradiation and PQQ treatment Furthermore, SIRT6 was shown to be crucial in the regulation were performed as described above. At 72 h post-irradiation, of mammalian longevity; therefore, it was hypothesized that total RNA was extracted from the cells using TRIzol reagent SIRT6 may have potential for the treatment of age‑associated (Promega Corp., Madison, WI, USA). RNA concentration and diseases (21,22). purity were determined using a Nanodrop‑2000 UV spectro- The present study aimed to determine whether PQQ photometer (Thermo Fisher Scientific). Ribosomal RNA band influences the damage of UVA irradiation, and how it func- integrity was evaluated using conventional denaturing agarose tions in the cell. It was hypothesized that PQQ protects gel electrophoresis (23) using the SDS-PAGE gel quick prepa- human dermal fibroblasts from senescence caused by ration kit (Beyotime Institute of Biotechnology, Shanghai, UVA irradiation. An in vitro cell‑senescence model was China). constructed through the exposure of PQQ‑pre-treated HDFs to UVA, and the effect of PQQ on protein and/or mRNA Reverse transcription quantitative PCR (RT‑qPCR). Samples expression levels of the senescence marker genes matrix were normalized to total RNA content prior to reference gene metalloproteinase (MMP)1 and MMP3 as well as SIRT1, assessment and gene expression analysis. Equal amounts of SIRT6, nuclear factor erythroid 2‑related factor 2 (Nrf2) and R NA (30 0 ng) from each sample were reverse‑transcribed using heme oxygenase 1 (HO‑1) were detected using polymerase a PrimeScript™ RT Reagent kit with gDNA Eraser (Takara chain reaction (PCR) and western blot analyses. Furthermore, Bio, Inc, Dalian, China) according to manufacturer's instruc- senescence‑associated β‑galactosidase (SA‑β‑Gal) staining tions. qPCR was performed using SYBR green dye method was used to determine the senescence status of HDFs (8). (SYBR Premix Ex Taq; Takara Bio Inc.) using an ABI700 SA‑β‑Gal activity distinguishes senescent cells from those Real‑Time PCR detection system (Applied Biosystems; Life which are quiescent or terminally differentiated, therefore Technologies, Thermo Fisher Scientific). Three biological and acting as a senescence biomarker. two technical replicates per biological sample were performed. The following standard cycling conditions for qPCR runs were Materials and methods applied: 95˚C for 3 min to activate polymerase, 40 cycles of denaturation at 95˚C for 15 sec and annealing‑extension Cell culture. Primary HDFs were purchased from at 60˚C for 30 sec. Melting curve analysis was performed BioHermes Bio& Medical Technology Co., Ltd (Wuxi, following every run by defined heating up to 95˚C to assess the China). Fibroblasts were subsequently cultured in a 10-cm presence of unspecific PCR products. A negative control was dish in Dulbecco's modifed Eagle's medium (DMEM; Gibco included in each assay run, using water instead of template. Life Technologies, Carlsbad, CA, USA) supplemented with Specific primers for the RT‑qPCR reaction were as 15% fetal bovine serum (FBS; Gibco Life Technologies), follows: MMP‑1 forward, 5'‑TTGGAGGGGATGCTCATT‑3' penicillin (100 U/ml) and streptomycin (100 µg/ml; Gibco and reverse, 5'‑ACACGCTTTGGGGTTTG‑3' with a product Life Technologies). Cells were divided into aliquots and size of 103 bp; MMP3 forward, 5'‑GCAGTTTGATCAGCC seeded into a six‑well plate (Corning Incorporated, Corning, TATCC‑3' and reverse, 5'‑TCCAGAGTGTCGGAGTCCAG‑3' N Y, USA), then incubated in an incubator (3111; Thermo with a product size of 138 bp; SIRT1 forward, 5'‑TGACTGTGA

Fisher Scientific, Waltham, MA, USA) at 37˚C with 5% CO2 AGCTGT ACG AGGAG‑3' and reverse, 5'‑GGA AGA CCC AAT in a humidified atmosphere. For cell culture, the medium AACAATGAGGA‑3' with a product size of 120 bp; SIRT6 was replaced every 2‑3 days and fibroblasts were passaged at forward, 5'‑CCCACGGAG TCT GGA CCAT‑3' and reverse, 80% confluence. Cells were then seeded in a 6-well plate at a 5'‑CTCTGCCAGTTTGTCCCTG‑3' with a product of 194 bp; density of 1x106 cell/cm2 for use in subsequent experiments. β-actin forward, 5'-TGGAATCTTGCTCTTATTTTCACA-3' and reverse, 5'-TAAAACGCAGCTCAGTAACAGTCCG-3'. UVA irradiation. At 24 h following the addition of PQQ All primers were synthesized by Sangon Biotech, Co., (Sigma‑Aldrich, St. Louis, MO, USA) into the culture media, Ltd. (Shanghai, China) and were used at 400 nM except for cells in the six‑well plate were exposed to 9 J/cm2 U VA i r radia- β‑actin at 300 nM. All PCR efficiencies were between 90 and tion. Cells were washed with phosphate-buffered saline (PBS; 110%. Expression data of reference genes were validated using Gibco Life Technologies) and covered with a thin layer of PBS geNorm 3.5 software (http://medgen.ugent.be/genorm/), as prior to UVA exposure. The culture plate lid was removed, and previously described (24). 4384 MOLECULAR MEDICINE REPORTS 12: 4382-4388, 2015

A B

Figure 1. β‑Galactosidase expression in HDFs following treatment with UVA and/or PQQ. (A) Representative images of HDFs treated as follows: a) Untreated, b) 50 ng/ml PQQ, c) 9 J/cm2 UVA irradiation, d) 50 ng/ml PQQ prior to 9 J/cm2 UVA irradiation, e) 100 ng/ml PQQ prior to 9 J/cm2 UVA irradiation (magnification, x20). Staining was performed 72 h after irradiation. (B) Quantitative analysis of the percentage of cells positive for β‑galactosidase activity following staining with senescence‑associated‑β‑galactosidase. Values are presented as the mean ± standard deviation (n=3). *P<0.05, compared with HDFs treated with UVA only. HDFs, human dermal fibroblasts; UVA, ultraviolet A; PQQ, pyrroloquinoline quinine.

A B

Figure 2. PQQ reduces the expression of senescence marker genes MMP1 and MMP3. Human dermal fibroblasts were irradiated with 9 J/cm2 UVA with or without pretreatment of 50 or 100 ng/ml PQQ and reverse transcription quantitative polymerase chain reaction was used to detect mRNA levels of (A) MMP1 and (B) MMP3. Total RNA was extracted 72 h after irradiation. Values are presented as the mean ± standard deviation (n=3). *P<0.05. PQQ, pyrroloquinoline quinine; UVA, ultraviolet A; MMP, matrix metalloproteinase.

SA‑β‑Gal staining. SA‑β‑Gal activity was evaluated no. SC-365949; BD Biosciences, San Jose, CA, USA), HO‑1 using a β‑galactosidase staining kit (Beyotime Institute of (mouse monoclonal IgG; cat. no. 610712; BD Biosciences) Biotechnology, Haimen, China). Cells were washed with PBS and GAPDH (mouse monoclonal IgG; cat. no. G8795; and then fixed for 15 min at room temperature with fixative Sigma-Aldrich), at a dilution of 1:1,000 and at 4˚C for 8 h. solution. The HDF cells were then incubated at 37˚C over- Horseradish peroxidase‑conjugated goat anti‑mouse and night. SA‑β‑Gal-positive staining is expressed as a percentage anti‑rabbit immunoglobulin G were used as secondary anti- of the total number of cells; cell numbers were counted in four bodies at a dilution of 1:5,000 (cat. no. A3682; Sigma‑Aldrich) continuous visual fields using a microscope Olympus( CX51; at room temperature for 2 h. GAPDH was used as an internal Olympus, Tokyo, Japan; total magnification, x20). control. Blots were analyzed using an Super Signal West Pico Enhanced Chemiluminisence western blot analysis Western blot analysis. Whole-cell lysate extracts were system (Thermo Fisher Scientific). Experiments were separated using 12% SDS‑PAGE, transferred to polyvi- performed three times under identical conditions. nylidene difluoride membranes (Roche Diagnostics, San Francisco, CA, USA), blocked with 5% bovine serum Statistical analysis. All statistical analyses were performed albumin (Sigma‑Aldrich) and detected using the following using GraphPad Prism software (GraphPad Inc., La Jolla, CA, primary antibodies: SIRT1 (mouse monoclonal IgG; cat. USA). Values are presented as the mean ± standard deviation. no. SC-74504; Santa Cruz Biotechnology, Inc., Dallas, TX, The one‑way analysis of variance was used for comparisons USA), SIRT6 (rabbit monoclonal IgG; cat. no. S4322; Santa involving more than two groups. P<0.05 was considered to Cruz Biotechnology, Inc.), Nrf2 (rabbit monoclonal IgG; cat. indicate a statistically significant difference between values. ZHANG et al: EFFECT OF PQQ ON HDF SENESCENCE FROM UVA IRRADIATION AND THE ANTI-APOPTOTIC PATHWAY 4385

Figure 3. Low-dosage PQQ increases the expression of SIRT1 and SIRT6. (A) mRNA and (B) protein expression levels of SIRT1 and SIRT6 in human dermal fibroblasts irradiated with 9 J/cm2 UVA with or without pretreatment with 50 or 100 ng/ml PQQ. GADPH was used as an internal control. Protein was extracted 72 h after the UVA irradiation. Values are presented as the mean ± standard deviation (n=3). *P<0.05. PQQ, pyrroloquinoline quinine; UVA, ultraviolet A; SIRT, sirtuin.

Figure 4. Expression pattern of SIRT1 and SIRT6 in UVA‑irradiated HDFs pretreated with PQQ over seven days. (A) mRNA and (B) protein expression levels of SIRT1 and SIRT6 in HDFs irradiated with 9 J/cm2 UVA with or without pre-treatment of 50 or 100 ng/ml PQQ were recorded daily for seven days. GADPH was used as an internal control. Values are presented as the mean ± standard deviation (n=3). *P<0.05. Total RNA and protein were extracted 72 h after the UVA irradiation. PQQ, pyrroloquinoline quinine; SIRT, sirtuin; UVA, ultraviolet A; HDFs, human dermal fibroblasts.

Results while 50 ng/ml PQQ attenuated the ratio of positive staining compared with that of the UVA‑only cells (29 vs. 53%, respec- Low-dose PQQ reduces SA‑β‑Gal activity in UVA‑induced tively; P<0.01) (Fig. 1A and B). In addition, the percentage of senescent HDFs. HDFs were pre-treated with various concen- positively stained cells following treatment with 100 ng/ml trations of PQQ (50 and 100 ng/ml) and then subjected to PQQ and UVA were comparable to those of the cells treated UVA irradiation in order to induce senescence, which was with 50 ng/ml PQQ and UVA (Fig. 1A and B). detected using an SA‑β‑Gal staining kit. The results showed that the percentage of cells stained by X‑gal following 9 J/cm2 Low-dose PQQ reduces the expression of senescence markers UVA irradiation was markedly increased compared with MMP1 and MMP3 in UVA‑irradiated HDFs. RT‑qPCR that in the control group (53 and 8%, respectively; P<0.05), analysis showed that mRNA expression levels of MMP1 and 4386 MOLECULAR MEDICINE REPORTS 12: 4382-4388, 2015

Figure 5. Low-dosage PQQ affects the expression of Nrf2 and HO‑1 in UVA‑irradiated HDFs. (A) mRNA and (B) protein expression of Nrf2 and HO‑1 in HDFs irradiated with 9 J/cm2 UVA with or without pre-treatment with 50 or 100 ng/ml PQQ. PQQ, pyrroloquinoline quinine; Nrf2, nuclear factor‑derived erythroid 2‑related factor 2; HO‑1, heme oxygenase 1; UVA, ultraviolet A; HDFs, human dermal fibroblasts.

Figure 6. Expression pattern of Nrf2 and HO‑1 in UVA‑irradiated HDFs pretreated with PQQ over seven days. (A) mRNA and (B) protein expression levels of Nrf2 and HO‑1 in HDFs irradiated with 9 J/cm2 UVA with or without pretreatment of 50 or 100 ng/ml PQQ were recorded daily for seven days. GADPH was used as an internal control. Values are presented as the mean ± standard deviation (n=3). PQQ, pyrroloquinoline quinine; Nrf2, nuclear factor‑derived erythroid 2‑related factor 2; HO‑1, heme oxygenase 1; UVA, ultraviolet A; HDFs, human dermal fibroblasts.

MMP3 were markedly reduced in UVA‑irradiated HDFs In order to further study the efficacy of PQQ, western blot pre-treated with 50 or 100 ng/ml PQQ as compared with those analysis was used to evaluate the effect of different concen- in the UVA‑only group (P<0.05) (Fig. 2). trations of PQQ on SIRT1 and SIRT1 protein expression in UVA‑irradiated HDFs. The results showed that 50 ng/ml and SIRT1 and SIRT6 expression levels are increased in 100 ng/ml PQQ had a significant effect on SIRT1 and SIRT6 UVA‑irradiated HDFs following pre-treatment with PQQ. At levels. Therefore, 50 ng/ml PQQ was used for the subsequent 72 h following the addition of 50 ng/ml PQQ to the culture evaluation of the effect of PQQ on UVA‑irradiation‑induced media of UVA-irradiated HDFs, mRNA and protein expression HDF senescence at numerous time‑points (0, 1, 2, 3, 4, 5, 6 levels of SIRT1 and SIRT6 were found to be increased (Fig. 3). and 7 days). The results showed that 50 ng/ml PQQ increased In addition, notable differences were observed in the protein mRNA and protein expression of SIRT1 and SIRT6 within levels of SIRT1 and SIRT6 at 72 h following UVA irradiation. 72 h. 50 ng/ml PQQ was added at the 5th day (Fig. 4). ZHANG et al: EFFECT OF PQQ ON HDF SENESCENCE FROM UVA IRRADIATION AND THE ANTI-APOPTOTIC PATHWAY 4387

PQQ may activate the Nrf2/HO‑1 pathway in UVA-irradiated protein expression of SIRT1, SIRT6, Nrf2 and HO‑1 following HDFs. As SIRT1/Nrf2/HO‑1 is a classical anti‑apoptotic U VA t reatment were partially recovered when treated with pathway (11,23,24), the present study investigated whether this 50 ng/ml PQQ. This observation indicated that Nrf2/HO‑1 pathway was involved in UVA‑induced cell apoptosis. Altered signaling may have an important role in the protective expression levels of Nrf2 and HO‑1 mRNA expression levels effect of PQQ against UVA irradiation‑induced senescence were not found to be significant in HDFs pre-treated with in HDFs. As the increased expression of SIRT1, Nrf2 and 50 ng/ml PQQ without UVA irradiation. However, following HO‑1 showed significant consistency, it was suggested that UVA irradiation, PQQ was shown to alter the mRNA and the SIRT1/Nrf2/HO‑1 pathway was activated following PQQ protein expression of Nrf2 and HO‑1 (Fig. 5A and B; Fig. 6). treatment in UVA‑irradiated HDFs. Of note, 50 ng/ml PQQ achieved the most significant effect on In conclusion, the results of the present study demonstrated the expression of Nrf2 and HO‑1 within 72 h. that PQQ protected HDFs from UVA irradiation in vitro, as shown by the upregulation of SIRT1 and SIRT6 following Discussion treatment with PQQ. In addition, the Nrf2/HO‑1 signaling pathway was activated. The present study provided experi- In a recent study, PQQ was reported to act as a growth factor mental evidence for the use of PQQ as a drug to prevent skin wh ich cont r ibuted to ma m ma l ia n g row t h a nd st i mulated epit he - cell senescence and aging. lial cells through the activation of epidermal growth factor receptor (25). PQQ has also been reported to protect primary Acknowledgements cultured hippocampal neurons against glutamate‑induced cell apoptosis, the mechanism of which was found to proceed The present study was supported by the Nanjing Medical via scavenging ROS and activating phosphatidylinositol Science and Technology Development Fund Project 3-kinase/Akt/glycogen sythase kinase 3β/Nrf2 signaling path- (grant no. 2011NJMU242). ways (26). In addition, PQQ was reported to attenuate oxidative stress‑induced cell damage in the heart, liver and brain (27). In References the present study, a photoaging model was constructed in order to study the protective effects of PQQ on UVA‑induced HDF 1. Kohl E, Steinbauer J, Landthaler M and Szeimies RM. Skin senescence as well as the mechanisms by which this proceeds. aging. J Eur Acad Dermatol 25: 873-884, 2011. The results demonstrated that low-dose PQQ protected HDFs 2. He YY, Council SE, Feng L and Chignell CF: UVA-induced cell cycle progression is mediated by a disintegrin and metal- from senescence following UVA irradiation, which may be loprotease/epidermal growth factor receptor/AKT/Cyclin D1 associated with the anti‑apoptotic Nrf2/HO‑1 and SIRT1 pathways in keratinocytes. Cancer Res 68: 3752-3758, 2008. pathways. 3. Krutmann J: Ultraviolet A radiation‑induced biological effects in human skin: relevance for photo aging and photodermatosis. Fibroblasts are the primary cellular components of the J Dermatol Sci 23 (Suppl 1): S22‑S26, 2000. human dermis and reflect the senescence status of skin at the 4. Salisbury SA, Forrest HS, Cruse WB and Kennard O: A novel cellular level (23). Long-wavelength UV radiation penetrates coenzyme from bacterial primary alcohol dehydrogenases. Nature 280: 843‑844, 1979. the epidermis to reach the dermis, where it exacerbates the 5. Felton LM and Anthony C: Biochemistry: role of PQQ as a photoaging of skin (21). In the present study, it was demonstrated mammalian enzyme cofactor? Nature 433: E10‑E12, 2005. that exposure of HDF to UVA under non‑toxic conditions 6. Matsushitak K, Toyama H, Yamada M, et al: Quinoproteins: structure function and biotechnologies applications. Appl increased the mRNA and protein levels of MMP1 and MMP3. Microboil Biot 58: 13‑22, 2002. In addition, the expression of SA‑β‑Gal, a biomarker of fibro- 7. Liu CL, Siesjö BK and Hu BR: Pathogenesis of hippocampal blasts in vivo and in vitro (8), was significantly upregulated in neuronal death after hypoxia‑ischemia changes during brain development. Neuroscience 127: 113‑123, 2004. UVA‑irradiated HDFs. However, PQQ reduced the expression 8. Dimri GP, Lee X, Basile G, et al. A biomarker that identifies of SA‑β‑Gal as well as mRNA expression levels of MMP1 and senescent human cells in culture and in aging skin in vivo. Cell MMP3. These findings suggested that PQQ has a protective Biology 92: 9363-9367, 1995. 9. Ohwada K, Takeda H, Yamazaki M, Isogai H, Nakano M, effect on cultured HDFs against UVA radiation. Shimomura M, Fukui K and Urano S: Pyrroloquinoline quinone Sirtuins (Sir) are class Ⅲ deacetylases, a family of proteins (PQQ) prevents cognitive deficit caused by oxidative stress in that require NAD as a co‑factor; therefore, they are controlled rats. J Clin Biochem Nutr 42: 29–34, 2008. 10. Zhang Q, Shen M, Ding M, Shen D and Ding F: The neuropro- by cellula r [NAD]/[NADH] ratios and respond to changes in the tective action of pyrroloquinoline against glutamate‑induced cellular metabolism (8), which is essential for lifespan exten- apoptosis in hippocampal neurons is mediated through the sion by calorie restriction in yeast, worms and flies (28). Seven activation of PI3K/Akt pathway. Toxicol Appl Pharm 252: 62‑72, 2011. Sir2 homologs (SIRT1‑7) have been identified in mammals 11. Kanfi Y, Naiman S, Amir G, et al: The sirtuin SIRT6 regulates and have important roles in the regulation of oxidative stress, lifespan in male mice. Nature 483: 218‑221, 2012. DNA damage and metabolism, therefore making them good 12. Hekimi S and Guarente L: Genetics and the specificity of the aging process. Science 299: 1351‑1354, 2003. candidates as lifespan regulators (29). PQQ has been shown 13. Howitz KT, Bitterman KJ, Cohen HY, et al: Small molecule to activate cyclic adenosine monophosphate response factors, activators of sirtuins extend Saccharomyces cerevisiae lifespan. including Nrf1, Nrf2 and mitochondrial transcription factors Nature 425: 191‑196, 2003. 14. Landry J, Sutton A, Tafrov ST, et al: The silencing protein SIR2 to enhance mitochondriogenesis (30-33). The present study and its homologs are NAD‑dependent protein deacetylases. Proc demonstrated that mRNA and protein expression of SIRT1, Natl Acad Sci USA 97: 5807‑5811, 2000. SIRT6 and Nrf2 were increased in a group of HDFs treated 15. Sugino T, Maruyama M, Tanno M, et al: Protein deacetylase SIRT1 in the cytoplasm promotes nerve growth factor‑induced with 50 ng/ml PQQ and UVA radiation; of note, improvements neurite outgrowth in PC12 cells. FEBS Lett 584: 2821‑2826, were observed within 72 h. In addition, decreased mRNA and 2010. 4388 MOLECULAR MEDICINE REPORTS 12: 4382-4388, 2015

16. Hisahara S, Chiba S, Matsumoto H, et al: Histone deacetylase 26. Vandesompele J, De Preter K, Pattyn F, et al: Accurate normal- SIRT1 modulates neuronal differentiation by its nuclear translo- ization of real‑time quantitative RT‑PCR data by geometric cation. Proc Natl Acad Sci USA 105: 15599‑15604, 2008. average of multiple internal control genes. Genome Biol 3: 17. Michán S, Li Y, Chou MM, et al: SIRT1 is essential for normal RESEARCH0034, 2002. cognitive function and synaptic plasticity. J Neurosci 30: 27. Li H, Wu S, Shi N, Lian S and Lin W: Nrf2/HO-1 pathway acti- 9695‑9707, 2010. vation by manganese is associated with reactive oxygen species 18. Jin Q, Yan T, Ge X, Sun C, Shi X and Zhai Q: Cytoplasm‑localized and ubiquitin-proteasome pathway, not MAPKs signaling. J Appl SIRT1 enhances apoptosis. J Cell Physiol 213: 88‑97, 2007. Toxicol 31: 690-697, 2011. 19. Byles V, Chmilewski LK, Wang J, et al: Aberrant cytoplasm local- 28. Kimura K, Takada M, Ishii T, et al: Pyrroloquinoline quinine ization and protein stability of SIRT1 is regulated by PI3K/IGF‑1R stimulates epithelial cell proliferation by activating epidermal signaling in human cancer cells. Int J Biol Sci 6: 599‑612, 2010. growth factor receptor through redox cycling. Free Radical Bio 20. Herskovits AZ and Guarente L: Sirtuin deacelylase in neurode- Med 53: 1239‑1251, 2012. genera‑ tive disease of aging. Cell Res 23: 746‑758, 2013. 29. Zhang Q, Ding M, Cao Z, et al: Pyrroloquinoline quinine protects 21. Herrmann G, Brenneisen P, Wlaschek M, et al: Psoralen photo- rat brain cortex against acute glutamate‑induced neurotoxicity. activation promotes morphological and functional changes in Neutovhrm Res 38: 1661‑1671, 2013. fibroblasts in vitro reminiscent of cellular senescence. J Cell 30. Li HH, He B, Peng H and Liu SQ: Effects of pyrroloquinoline Sci 111: 759‑767, 1998. quinine on proliferation and expression of c‑fos, c‑jun, CREB 22. Brugè F, Venditti E, Tiano L, Littarru GP and Damiani E: Reference and PCNA in cultured Schwann cells. Zhonghua Zheng Xing gene validation for qPCR on normoxia‑ and hypoxia‑cultured Wai Ke Za Zhi 27: 298‑303, 2011 (In Chinese). human dermal fibroblasts exposed to UVA: is β‑actin a reliable 31. Blander G and Guarente L: The Sir family of protein deaceylases. normalize for photoaging studies? J Biotechnol 156: 153‑162, 2011. Annu Rev Biochem 73: 417‑435, 2004. 23. Kawahara TL, Michishita E, Adler AS, et al: SIRT6 links histone 32. Lin SJ, Defossez PA and Guarente L: Requirement of NAD H3 lysine 9 deacetylation and organismal life span. Cell 136: and SIR2 for life‑span extension by calorie restriction in 62‑74, 2008. Sacharomyces cerevisiae. Science 289: 2126‑2128, 2000. 24. Herrmann G, Wlaschek M, Lange TS, et al: UVA irradiation stim- 33. Frye RA: Characterization of five human cDNAs with homology ulates the synthesis of various matrix‑metalloproteinases (MMPs) to the yeast SIR2 gene: Sir2‑like proteins (sirtuins) metabolize in cultured human fibroblasts. Exp Dermatol 2: 92‑97, 1993. NAD and may have protein ADP‑ribosyltransferase activity. 25. Masek T, Vopalensky V, Suchomelova P and Pospisek M: Biochem Biophys Res Commun 260: 273‑279, 1999. Denaturing RNA electrophoresis in TAE agrose gels. Anal Biochem 336: 46‑50, 2005.